Up to date
Qwen Apache 2.0 out 13 months ago
4.41B parameters, text in, text out, 32K tokens of context. 1 provider serves it.
Verified 2026-10-25 Rank #609 in LLMs · score 73.9 method v1 Sources 3 Downloads 30d 19.3k · #460 of 619 tracked
Picked by rule from 1 tracked piece: the lab's own post, the most shared, a field report, the latest.
Qwen3Guard introduces multilingual safety guardrail models featuring fine-grained tri-class judgments and real-time token-level monitoring for large language models.
Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.
| # | Cheapest live row | In /1M | Out /1M | Context | Quant | Seen |
|---|---|---|---|---|---|---|
| 1 | Featherless | — | — | — | — | Oct 25 |
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
Served by 1 API provider; 11 quantized builds for local use.
Weights only, an estimate: leave room for the KV cache. Method
About 2.6 GB at INT4, 5.3 GB at FP8, 11 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2025-09-23), amber triangles are 1 dated article (each links to it), grey ticks are repo commits.
28 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Everything else, folded: the full tables, every source.
Qwen3Guard-Gen-4B is a 4.41B parameter generative safety moderation model based on the Qwen3 architecture.[1][2] The model is released under the Apache 2.0 license and supports 119 languages and dialects.[3][4]
Generated 2026-10-07 15:13 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Listed in LLMs
| Pipeline tag | text-generation |
|---|---|
| Library | transformers |
| Architectures | Qwen3ForCausalLM |
| Model type | qwen3 |
| Dtype | bfloat16 |
| Params by dtype | BF16: 4,411,424,256 |
| Size band | M |
| MoE | no |
| Context length | 32,768 (config.json max_position_embeddings) |
| Papers | arXiv:2510.14276 |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · Qwen3ForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · Qwen3ForCausalLM · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Weights-only memory for Qwen3Guard-Gen-4B: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 11 GB · Single GPU (<24 GB) |
|---|---|
| FP8 (1 byte/param) | 5.3 GB · Single GPU (<24 GB) |
| INT4 (0.5 bytes/param) | 2.6 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
Base model: Qwen/Qwen3-4B
quantized: 11finetune: 9adapter: 8
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| mradermacher/Qwen3Guard-Gen-4B-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 3 | 6.7k |
| DevQuasar/Qwen.Qwen3Guard-Gen-4B-GGUF | quantized | GGUF | — | — | — | DevQuasar | 0 | 1.6k |
| geoffmunn/Qwen3Guard-Gen-4B-GGUF | quantized | GGUF | — | — | apache-2.0 | geoffmunn | 0 | 1.2k |
| mediterranean1973/Qwen3Guard-Gen-4B-awq-int4 | quantized | awq | 4.44B | M | apache-2.0 | mediterranean1973 | 0 | 952 |
| EddyLuo/FoodGuard-4B | finetune | — | 4.02B | M | apache-2.0 | EddyLuo | 2 | 555 |
| PLJIANGG05/qwen3guard-gen-4b-danger-intent-opendata-qwensynth-supplement-sampled4000-v9 | finetune | — | 4.02B | M | cc-by-nc-4.0 | PLJIANGG05 | 0 | 627 |
| i1see1you/VirbiusGuard-4B | quantized | GGUF | 4.02B | M | apache-2.0 | i1see1you | 1 | 497 |
| Yunhao-Feng/AdaGuard-4B | finetune | — | 4.41B | M | apache-2.0 | Yunhao-Feng | 0 | 477 |
| amalia-llm/amaliaguard-4b | finetune | — | 4.02B | M | — | amalia-llm | 0 | 456 |
| liodon-ai/Qwen3Guard-Gen-4B-imatrix-GGUF | quantized | imatrix | — | — | other | liodon-ai | 0 | 449 |
| zenlm/zen-guard | finetune | — | 4.41B | M | apache-2.0 | zenlm | 0 | 364 |
| JeremyLiShuhao/UD_model_v1 | finetune | — | 4.02B | M | apache-2.0 | JeremyLiShuhao | 0 | 275 |
| burrowswang/Qwen3Guard-Gen-4B-Q4_K_M-GGUF | quantized | GGUF | — | — | apache-2.0 | burrowswang | 0 | 256 |
| HikmaAI/HikmaGuard | finetune | — | 4.02B | M | apache-2.0 | HikmaAI | 3 | 54 |
| mistralrs-community/Qwen3Guard-Gen-4B-UQFF | quantized | — | — | — | — | mistralrs-community | 0 | 166 |
| AhmedxSaad/Qwen3Guard-Gen-4B-Q4_K_M-GGUF | quantized | GGUF | — | — | apache-2.0 | AhmedxSaad | 0 | 139 |
| abnormalmapstudio/Qwen3Guard-Gen-4B-mxfp4-mlx | quantized | mxfp4 | 4.02B | M | apache-2.0 | abnormalmapstudio | 0 | 129 |
| Otilde/Qwen3Guard-Gen-4B-MLX-Q8 | quantized | MLX | 4.02B | M | apache-2.0 | Otilde | 0 | 129 |
| Otilde/Qwen3Guard-Gen-4B-Heretic | finetune | — | 4.02B | M | apache-2.0 | Otilde | 0 | 85 |
| psssd-cat/HaloGuard1.2-4B | adapter | — | — | — | — | psssd-cat | 0 | 44 |
| psssd-cat/HaloGuard1.3-4B | adapter | — | — | — | — | psssd-cat | 0 | 43 |
| psssd-cat/HaloGuard1.5-4B | adapter | — | — | — | — | psssd-cat | 0 | 35 |
| psssd-cat/HaloGuard1.6-4B | adapter | — | — | — | — | psssd-cat | 0 | 34 |
| psssd-cat/HaloGuard1.1-4B | adapter | — | — | — | — | psssd-cat | 0 | 33 |
| psssd-cat/HaloGuard1.7-4B | adapter | — | — | — | — | psssd-cat | 0 | 33 |
| psssd-cat/HaloGuard1.4-4B | adapter | — | — | — | — | psssd-cat | 0 | 31 |
| PinoCookie/Qwen3Guard-Gen-4B-Abliterated | finetune | — | — | — | apache-2.0 | PinoCookie | 0 | 19 |
| TuanAnhHoangBui/qwen3guard-gen-4b-vi-en-p-pr-lora | adapter | — | — | — | apache-2.0 | TuanAnhHoangBui | 0 | 13 |
Hosted endpoints observed for Qwen3Guard-Gen-4B, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting.
| Provider | Model ref | Context | Quant | $/M in | $/M out | tok/s | TTFT | Tools | Status | Observed |
|---|---|---|---|---|---|---|---|---|---|---|
| Featherless hf | Qwen/Qwen3Guard-Gen-4B | — | — | — | — | — | — | — | live | 2026-10-25 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.
arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:41 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:13 UTC · Last verified 2026-10-25
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.
